Behavioral Ecology Vol. 13 No. 6: 728-736
© 2002 International Society for Behavioral Ecology
Structural coloration and sexual selection in the barn swallow Hirundo rustica
a Laboratoire d'Ecologie Evolutive Parasitaire, CNRS UMR 7103, Université Pierre et Marie Curie, 7 quai St. Bernard, Case 237, F-75252 Paris Cedex 05, France b Departamento de Biología Animal, Universidad de Extremadura, E-06071 Badajoz, Spain
Address correspondence to P. Ninni. E-mail: pninni{at}snv.jussieu.fr.
Received 1 December 2001; revised 15 January 2002; accepted 16 January 2002.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Structural coloration has been hypothesized to play a role in sexual selection, and we tested whether this was the case in a field study of the barn swallow Hirundo rustica. The dorsal iridescent plumage of barn swallows has a strong reflectance in the ultraviolet (UV) region, with adult males on average reflecting 8-9% more than adult females, as revealed by a 2-year study in southwestern Spain. The correlation between structural coloration (described by the reflectance in the UV part of the spectrum, UV chroma and blue chroma) and three other secondary sexual characters significantly associated with male mating success (tail length, tail asymmetry, and red facial coloration) was weak and generally nonsignificant. Nor was there a significant relationship between color parameters and body condition. We tested for an association between structural coloration of the dorsal plumage and sexual selection in a number of independent tests. Arrival date of males was not significantly related to color; there was no significant relationship between coloration and probability of survival or age; mated males did not have stronger reflectance than unmated males; and the duration of the premating period was not significantly related to color. Reproductive success was not significantly correlated with plumage coloration in males, nor was the feeding rate of offspring by brightly colored males higher than that of males with less bright plumage. Given that sample sizes were large, and the power of statistical tests high, we conclude that current sexual selection on the coloration of the dorsal plumage in the barn swallow is, at best, weak.
Key words: barn swallows, condition dependence, feeding rate, Hirundo rustica, mating success, multiple signals, structural coloration, survival, ultraviolet spectrum.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Animals generally produce multiple sexual signals that differ in their phenotypic basis. Some signals are morphological representations of size and symmetry, others are colors based on carotenoids, melanins, or structural color. Why such a diversity exists, even in the same individuals, is a matter of great theoretical and empirical interest (Badyaev et al., 2001
;
Møller and Pomiankowski,
1993). Here we analyzed a signal based on structural color that
coexists with a number of other signals based on morphology, melanins, and
carotenoids.
The visual system of birds, which is based on four spectrally distinct,
single cone types associated with colored oil droplets
(Bowmaker et al., 1997),
forces any study carried out on aspects of avian behavior based on vision to
consider the reflectance of light and analyze spectral features in the entire
spectrum of sensitivity (300-700 nm;
Bennett et al., 1994). The
avian visual system in about 30 species is able to detect wavelengths down to
approximately 320 nm, which includes the near-ultraviolet light (320-400 nm;
Cuthill et al., 2000). This
particular light range, which is not perceived by humans, seems to play a role
in avian communication, and it is used by birds in orientation
(Able and Able, 1993;
Jacobs, 1992), in feeding, and
in searching for food (Church et al.,
1998, Siitari et al.,
1999).
Several studies have dealt with the role of ultraviolet (UV) plumage
reflectance in sexual signaling and mate choice, and sexual dimorphism has
been demonstrated in different species
(Andersson and Amundsen, 1997;
Andersson et al., 1998; Bennett
et al., 1996,
1997; Hunt et al.,
1997,
1998,
1999;
Johnsen et al., 1998).
Furthermore, UV reflectance reveals male age in bluethroats Luscinia
svecica (Andersson and Amundsen,
1997), and a study of blue grosbeaks Guiraca caerulea
indicated that UV reflectance is condition dependent
(Keyser and Hill, 1999),
reliably reflecting individual phenotypic quality
(Keyser and Hill, 2000).
Whether the UV spectrum represents a special and private channel for avian
signaling compared to the rest of the visual spectrum still remains to be
demonstrated (Banks, 2001),
although a recent study could not find a specific preference for the
reflectance in the UV compared to other parts of the reflectance spectrum
(Hunt et al., 2001). In the
present study, we investigated dorsal plumage structural coloration in the
barn swallow Hirundo rustica, in which the ability to perceive UV
light has been demonstrated (Chen and
Goldsmith, 1986; Chen et al.,
1984). The barn swallow has a dorsal plumage that appears black
and iridescent to the human eye, giving a bright blue tinge to the feathers
when observed under a particular angle of light.
The barn swallow is an ideal species for investigating the functional
significance of structural coloration and reflectance in the UV because a
great deal is known about sexual selection on several other morphological and
behavioral characters. Long-tailed males are preferred as social and sexual
mates (Møller, 1988,
1994b;
Saino et al., 1997), as are
males with symmetric outermost tail feathers (Møller,
1992,
1993,
1994c) and males with more
intense red facial plumage (Ninni et al., in preparation). The expression of
these secondary sexual characters in male barn swallows is condition dependent
(Camplani et al., 1999;
Møller, 1989,
1994b;
Saino et al., 1999), and males
with more exaggerated traits have a survival advantage (Møller,
1989,
1994b). Males with long tails
arrive early from the African winter quarters
(Møller, 1994a), and
they mate early, with a resultant early start of breeding and large seasonal
reproductive success (Møller,
1988,
1994a,b).
Attractive males provide absolutely and relatively little food to their
offspring (de Lope and Møller,
1993; Møller,
1994b,d).
The dorsal feathers of a male are clearly visible to a female during the
aerial display before mating, since males fly slowly in front of the female
while singing on the way to a potential breeding site (see full description in
Møller, 1994b), and the
dorsal feathers may potentially be used as a cue in mate choice and sexual
signaling. This information allows determination of the relationship between
structural coloration and the expression of additional characters repeatedly
shown to be associated with sexual selection in the barn swallow.
First, we investigated whether there was sexual dimorphism in reflectance and spectral parameters of the dorsal feathers. Second, we determined whether individuals were consistent in coloration among years and whether this was associated with a selective advantage in terms of survival. Third, we determined whether coloration was positively correlated with the expression of three other secondary sexual characters (length and symmetry of the outermost tail feathers, the intensity of the red facial plumage) and body condition in male barn swallows. Fourth, we determined whether coloration was correlated with male arrival date, male mating success, duration of the prelaying period for males, male reproductive success, and absolute feeding rate of offspring by males, characters that are quality indicators of individual barn swallows.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Study population
We studied barn swallows at Badajoz (38°50' N, 6°59' W), southwestern Spain, during the reproductive seasons 1999-2000, as part of a long-term project. The study area consists of open farmland with pastures, cereals and fruit plantations, and almost all barn swallows breed in barns (de Lope and Møller, 1993
).
Capture and ringing
We captured adults once a week by closing all windows and doors of
buildings with breeding barn swallows. A single door was opened the next
morning, and birds were then captured in a mist net. All birds were provided
with an individually numbered aluminium ring and a combination of color
plastic rings. Before releasing the bird, we applied a unique combination of
colors on the white ventral feathers to allow easy identification in the
field. Birds were assumed to be yearlings when first captured without a ring,
while older birds already had the numbered aluminium ring from a previous
year. This assignment of age to birds was reliable as demonstrated by all
local recruits (more than 150) that had been ringed in a nest indeed were
captured the first time as adults when yearlings. Because adults always return
to their previous breeding site in subsequent years (as shown by none of more
than 3000 birds ever moving to another farm between years), birds that were
not recaptured in a year were considered to be dead. This assumption was
supported by the fact that no birds were missing in one year but recaptured in
a subsequent year. We sexed all birds from the presence of a brood patch (only
females) and the shape of the cloacal protuberance (males have larger
protuberances).
We captured 195 (102 males and 93 females) and 181 (100 males and 81 females) individuals in the two field seasons, respectively. We recaptured 74 individuals in the second field season that were already present the first year. When data from both years were pooled, we randomly chose data for an individual from only one year to assure that each individual only appeared once in the analyses, reducing the number of birds to 161 in 1999 and 141 in 2000 in the pooled analyses.
Morphological measurements
Upon capture each individual was placed in a dark bird cage and measured in
a standardized way to record the following variables (only some of these were
used here): flattened right and left wing length, length of the right and left
outermost tail feathers, length of the central tail feathers, and wing span to
the nearest millimeter using a ruler; beak length to feathering, beak depth at
the distal end of the nares, beak width at the commissure, tarsus length and
keel length with a digital calliper to the nearest 0.01 mm; and body mass with
a Pesola spring balance to the nearest 0.1 g. We took a blood sample (about
200 µl from ulnar vain) for genetic and hematological studies, and after
marking, released the bird. We calculated tail length as the mean of the
length of left and right outermost tail feathers and calculated tail asymmetry
as the absolute value of the difference between the length of the left and the
right outermost tail feathers. Body condition was calculated as the ratio
between the weight at capture and cube root tarsus length. Every adult
captured had four feather samples of three to six feathers plucked from the
center of the red throat badge, the center of the red forehead, the center of
the back from a site between the tips of the secondaries, and the center of
the ventral part opposite to the dorsal site of sampling. All feathers were
placed in a small plastic bag that was sealed and numbered with the ring
number of the bird and the date. This bag was subsequently stored in an
envelope in total darkness at room temperature until measurements were made.
In this study we used the dorsal feathers and the red throat feathers. Birds
recaptured in successive capture sessions and already marked and measured were
immediately released.
Arrival and breeding parameters
Arrival date of an individual was determined as the first capture date.
Although we used an interval of 1 week between successive capture sessions, we
considered the first capture date to be the arrival date of that individual
because birds arrived over a period of at least 2 months in both years. We
recorded the breeding territory of each individual from regular observations
(at least twice a week) from a hide at the breeding sites. We determined the
pairing status of males from the presence of a female next to the male and the
obvious behavior of unmated males (intense singing and sexual displays and
chases of approaching females, including neighboring females). The duration of
the premating period was determined as the interval (in days) between the date
of the first capture of an individual male and the date of laying. We visited
nests at least weekly, and more frequently during laying, hatching, and
fledging, to determine start of laying, total number of eggs laid, and total
number of fledglings produced per pair during each clutch and the entire
breeding season. During the nestling period, male and female feeding rate was
observed for 1 h in the morning during at least 7 days from days 3 to 14,
counting the number of times each mate visited the nest and fed the nestlings.
We made observations with binoculars from a hide positioned next to the focal
nest to allow identification. We calculated absolute feeding rate as the
average value per hour over the entire nestling period divided by the number
of nestlings in the nest. We tested for a correlation between plumage
coloration and reproductive success in a number of different ways: number of
eggs laid by the female during the first clutch (data available for both
reproductive seasons), number of eggs laid in the season, number of fledglings
in the season, and reproductive success calculated as the proportion of
fledglings relative to the total number of eggs laid by females. These last
three variables were only available for the second field season because an
independent experiment carried out in 1999 prevented us from using
reproductive success for the entire breeding season.
Color measurements
We analyzed plumage coloration (both dorsal and throat feathers) using a
spectroradiometer (Ocean Optics Europe), and measurements were made in the
range 300-700 nm and 400-700 nm for dorsal and throat feathers, respectively,
under controlled laboratory conditions. We used the range 400-700 nm for the
throat feathers because we only considered light absorbance as due to lutein
at the wavelength around 450 nm, and no reflection in UV region was detected
in preliminary measurements.
The feathers were subjected to a light source (DH2000, Top Sensor System) with a spectral range from 280 to 800 nm. They were placed under a metal chamber provided with a magnifying lens pointing at the part of the feather to be illuminated by the light source, which allowed, with a black cover, elimination of ambient light during measurement once the feather was placed correctly under the light spot. Two quartz optic fibers (Ocean Optics) were fixed at 90° and 45° to the surface of the sample on the metal chamber. The first one transferred the light (that thus reached the sample perpendicularly) from the source lamp to the feather, and the second one collected the reflected light and transferred it to the spectrometer (S2000). A DAQ Card 700 converted the data collected by the spectrometer into digital information and passed them on to a computer where software (Spectrawin 3.1) calculated reflectance data relative to a white reference tile (WS-2) and to the dark from 300 to 700 nm. Both white and dark references were recorded under the same metal chamber before a new session of measurements started, and they were checked regularly during measurements. Five measures (each being an average of 10 scans) were taken from each feather to be measured. Measurements were made on approximately the same area of the feather, moving the metal chamber entirely between two measures. The area of the visible surface of the feather that was measured was about 1 mm wide and situated at about 1 mm from the distal end of the feather. After repeatability analyses, we measured one feather for each individual, randomly chosen among those collected in the field, as described. All measurements were made by the same person (C.P. and P.N. for dorsal and throat feathers, respectively), and they were done blindly with respect to the identity of individuals.
The software Spectrawin gives a value for the black and white references and the sample reflectance data at each 0.4 nm interval (n) between 300 and 700 nm. Percentage reflectance at each n point could be calculated with a software by the formula: R(n) = {[(sample (n) - dark (n)]/[white (n) - dark (n]} x 100. From these data we calculated total reflectance (sum of reflectance data in the interval 300-700 nm), UV reflectance (sum of reflectance data in the interval 300-400 nm), blue reflectance (sum of reflectance data in the interval 400-475 nm), UV chroma (UV reflectance/total reflectance [R300-400/R300-700]), blue chroma (blue reflectance/total reflectance [R400-475/R300-700]), and hue (wavelength where maximal reflectance occurred).
Because the color of the red throat feathers is partially determined by
carotenoids, in particular lutein (Saino
et al., 1999; Stradi,
1998), which peaks at 450 nm, we calculated the mean reflectance
of light of these feathers in the interval 445-455 nm. From this value we
calculated light absorbance, as due to carotenoid content in the feathers, as
—log(reflectance) (for detailed description of procedure and
repeatability analyses, see Saino et al.,
1999). This value was used to test for a correlation between
structural color and carotenoid-based color.
Repeatability estimates
We estimated the reliability of our data by making repeated measurements to
obtain estimates of repeatability
(Falconer and Mackay, 1996).
Feathers from all individuals were measured five times. During these
measurements the optic fiber was removed and replaced on the feather in a
standardized way between each measurement as described above. Repeatability
within these measurements was higher than 80% for all color parameters (total
reflectance: R = 0.85, F = 29.49, df = 180,724, p
< .0001; UV reflectance: R = 0.85, F = 30.09, df =
180,724, p < .0001; blue reflectance: R = 0.87,
F = 33.62, df = 180,724, p < .0001; UV chroma: R
= 0.89, F = 41.94, df = 180,724, p < .0001; blue chroma:
R = 0.91, F = 50.93, df = 180,724, p < .0001;
hue: R = 0.86, F = 32.77, df = 180,724, p <
.0001). We measured 40 individuals twice. In this case the estimate of
repeatability was intermediate and higher when all the 10 measurements were
considered in the analysis (total reflectance: R = 0.49, F =
10.51, df = 39,360, p < .0001; UV reflectance: R = 0.42,
F = 8.32, df = 39,360, p < 0.001; blue reflectance:
R = 0.48, F = 10.33, df = 39,360, p < .0001; UV
chroma: R = 0.68, F = 22.56, df = 39,360, p <
.0001; blue chroma: R = 0.52, F = 11.65, df = 39,360,
p < .0001, hue: R = 0.61, F = 16.74, df =
39,360, p < .001).
Statistical methods
We tested the frequency distributions for normality and used parametric or
nonparametric tests accordingly. In our test for a relationship between dorsal
plumage coloration and mating success we matched unmated males and a randomly
chosen mated male that arrived on the same date in the same colony.
Differences in phenotype between these two categories of males were examined
using paired t tests. We tested for factors influencing the
probability of survival in logistic regression models with the binary survival
variable as the dependent variable, and sex, plumage parameters, and their
interactions as independent variables. To test for a correlation between color
variables and premating period, we used the Kendall partial correlation
coefficient to control for individual arrival date, which is strongly
positively correlated with laying date in barn swallows
(Møller, 1994b). All
tests were two tailed. We applied the Bonferroni correction to reduce the
number of cases with significance arising by chance because of multiple tests.
To avoid any problems of error rates, we followed Chandler's
(1995) recommendation of using
a significance level of 10%.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Descriptors of structural coloration
In Figure 1 we show two examples of reflectance spectra recorded from the dorsal feathers of barn swallows. The variability in the measurements was high in the amplitude of the peak of maximum reflectance (which strongly correlated with UV reflectance; both years of study, rs > .90), that ranged from 20% to 95% of the white reference. Thirty percent of the individuals (N = 122) measured in the 2 years presented a peak of maximum reflectance below 400 nm.
|
We calculated the Spearman rank correlation between the color parameters and applied the Bonferroni correction (Table 1). The correlation matrix revealed that total reflectance, and blue and UV reflectance were strongly positively correlated in both years. Blue and UV chroma were correlated with the reflectance in their part of the spectrum in both years. Hue described as the wavelength where the maximum reflectance occurs was strongly negatively correlated with UV chroma, less strongly with UV reflectance, and positively with blue chroma.
|
To describe the color of the dorsal plumage, we retained three variables that best accounted for the spectrum and allowed us to reduce the number of tests performed: UV reflectance (which describes both total and blue reflectance), UV chroma (a measure of hue), and blue chroma. Our decision was verified performing a principal components analysis using the varimax procedure on all color variables. We describe the results obtained with data from 1999, which were qualitatively similar to those obtained in 2000. The first component (PC1) accounted for 48% of the variation in the original spectra. The coefficient that relates PC1 to the original variables was 0.40 for the three variables describing reflectance (UV, blue, and total reflectance). The second component accounted for a further 43% of the variation, and the coefficients were 0.56 and negative for hue and 0.56 and positive for UV chroma. Therefore, this factor clearly described the shift of spectra to the ultraviolet part. The third component whose coefficient was strongly positive related to blue chroma (0.80) explained 8% of the variation. The three principal components accounted for 99% of the variation in both years. Based on the results of the principal components analysis and the correlation analyses, we subsequently used UV reflectance to describe the general reflectance of the spectrum, UV chroma which is also a measure of hue, and blue chroma to quantify the relative importance of the visible blue part of the spectrum.
Repeatability of reflectance between years
We found no significant repeatability between years for reflectance in the
ultraviolet part of the spectrum, nor for blue chroma (UV reflectance:
F = 0.90, df = 73,74, p = .68; blue chroma: F =
1.11, df = 73,74, p = .33). UV chroma showed a trend toward
consistency between years (F = 1.48, df = 73,74, p = .05),
although the repeatability was low (R = 0.19).
Sexual dichromatism
Mean values of reflectance in the UV were slightly different between years
(F = 3.76, df = 1,300, p = .053), with a larger value in
1999 than in 2000 (mean ± SE; 1999: 5231 ± 125, n =
195; 2000: 4800 ± 95.9, n = 181). A greater difference was
found in UV chroma (F = 17.50, df = 1,300, p < .0001;
1999: 0.316 ± 0.003, n = 195; 2000: 0.293 ± 0.003,
n = 181), whereas blue chroma did not show any clear variation
between years (F = 2.48, df = 1,300, p = .12; 1999: 0.259
± 0.001, n = 195; 2000: 0.264 ± 0.001, n =
181).
To test for a sexual difference in color, we added sex as a factor to the previous model and its interaction with year. Significant sexual dichromatism was found in UV reflectance of the plumage and in blue chroma, but no significant difference was found in UV chroma (Table 2). Males had a higher total reflectance in UV and higher values of blue chroma than females in both years (Table 3). In no case did we find a significant interaction between sexual dichromatism and year.
|
|
Correlations between reflectance and quality indicators
The correlations between tail length and measures of reflectance were
generally weak. In males no correlation was statistically significant
(Table 4). Tail symmetry was
not significantly correlated with coloration in either year
(Table 4), nor was there a
significant correlation with light absorbance of the red throat feathers
(Table 4). Finally, reflectance
was not significantly related to body condition
(Tables 4). In general, after
Bonferroni correction, there was no significant correlation between
condition-dependent characters and reflectance, and the same applied to body
condition.
|
Sexual selection and reflectance
Early arrival is associated with male mating success, and hence the
expression of secondary sexual characters, in the barn swallow
(Møller, 1994a). Male
arrival date was not significantly correlated with UV reflectance or blue or
UV chroma in any year after Bonferroni correction
(Table 5).
|
Male mating success was not significantly correlated with plumage reflectance. UV reflectance did not differ significantly between mated and unmated males (UV reflectance, 1999: t = 1.63, df = 12, p = 0.129; 2000: t = -1.39, df = 12, p = .190; UV chroma 1999: t = -0.390, df = 12, p = .70; 2000: t = -0.456, df = 12, p = .65). This was also the case for blue chroma (1999: t = 0.029, df = 12, p = .98; 2000: t = 0.393, df = 12, p = .70). Duration of the premating period was not significantly related to reflectance (Table 5), nor was there a significant correlation between male feeding rate and plumage coloration (Table 5).
Reproductive success of males, measured as the number of eggs laid by their mates in the first clutch and in the entire season, was not significantly correlated with color variables (Table 5). Reproductive success measured as the number of fledglings in the breeding season or the proportion of fledglings from the eggs laid showed a negative correlation with UV reflectance and UV chroma, but no relationship was found with blue chroma (Table 5).
Reflectance, survival, and age
We tested whether plumage reflectance was related to survival while taking
sex into consideration using a logistic approach. None of the colorimetric
variables predicted probability of survival (results not shown).
Yearlings did not show differences in plumage coloration compared to adults, while controlling for sex and year of sampling, nor was there evidence of interactions among age, sex, and year (Table 6; effect of age: UV reflectance: F = 1.27, df = 1,298, p = .261; UV chroma: F = 2.82, df = 1,298, p = .094; blue chroma: F = 2.09, df = 1,298, p = .149). We note that, in the model, sex and year effects showed the same patterns of difference as in previous analyses (results not shown): sexual dichromatism in UV reflectance and blue chroma seem indepdent of age in the barn swallow.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Sexual dichromatism
We analyzed color parameters of the iridescent dorsal plumage of barn swallows in relation to sex, age, and mating and reproductive success during two field seasons. We did not find any evidence that this character is under current sexual selection. We note that our measurement errors when measuring feathers in a standardized way under laboratory conditions were relatively large. However, we have been unable to locate other estimates of measurement error in studies of UV reflectance in the literature, taken on live birds or on feathers measured in the laboratory. This is worrisome because measurement errors must be quantified before any conclusions can be made. In addition, measurement errors assessed from blind tests will allow quantification without bias. Again, we have found no previous studies making blind tests. It seems unlikely that measurement errors in the laboratory are greater than measurement errors on live birds. Measuring plumage coloration in the field is not always possible for logistic and animal welfare reasons. In particular, in barn swallows, the number of measurements already taken on live birds (as part of a larger long-term project) and the number of individuals captured each session (which can range from few up to 100 birds) does not allow systematic measurement of the plumage at each capture session. Moreover, these factors will also preclude analyses of repeatability. In laboratory analyses, environmental and light conditions can be controlled, and feathers can be placed in a standard position with the light reaching the sample at a standard angle. Furthermore, the surface measured is always the same, and the background is standard. Nevertheless, we consider that the repeatability analyses and the high number of individuals studied justify our conclusions.
The dorsal plumage of the barn swallow is sexually dichromatic by 8-9% in
terms of reflectance in the ultraviolet spectrum (300-400 nm), and by 3% for
blue chroma (R400-475/R300-700),
whereas chroma in the UV part of the spectrum
(R300-400/R300-700) did not show any
sex difference (Tables 2 and
3). Sexual dimorphism in the
barn swallow is larger for the length of the outermost tail feathers,
averaging 10% in the present population
(Møller, 1995), and
sexual dimorphism is 15% for the length of white tail spots
(Kose and Møller, 1998)
and 5% for reflectance of red feathers of the throat patch (Ninni and
Møller, unpublished data). Thus, the degree of sexual dimorphism for
reflectance in the UV range is relatively large. The most common explanation
for the presence of sexual differences is sexual selection
(Andersson, 1994), with
selective advantages for one sex resulting in divergent selection on the
expression of the trait. Previous studies of birds have also shown various
degrees of sexual dichromatism in the UV spectrum (e.g.,
Andersson et al., 1998; Cuthill
et al., 1999,
2000) and differential mating
success or investment related to ultraviolet reflectance. The level of sexual
dichromatism in the barn swallow was slightly larger in 1999 than in 2000.
Such differences among years indicate that ecological factors before or during
the molt in the African winter quarters may affect reflectance of dorsal
feathers of adult barn swallows. The degree of sexual dichromatism was, in
fact, larger for UV in the year with larger mean values (1999).
Condition dependence of structural coloration
Many secondary sexual characters are condition dependent in their
expression, with males in prime condition enjoying a mating advantage because
of the possession of more extreme secondary sexual characters (review in
Johnstone, 1995). We have
investigated the evidence for condition dependence of plumage reflectance in
the barn swallow using three different tests: (1) correlations between
reflectance and other secondary sexual characters that are condition
dependent; (2) correlations between reflectance and body condition; and (3)
correlations between reflectance, survival and age.
First, we investigated whether the degree of plumage reflectance was
significantly positively associated with the expression of other secondary
sexual characters such as tail length, tail symmetry, and red facial
coloration. These traits have been shown in previous studies to be condition
dependent (Camplani et al.,
1999; Møller,
1989,
1994a,b;
Saino et al., 1999). Thus we
predicted that UV and blue coloration should be positively correlated with the
expression of these condition-dependent characters. None of the correlations
was statistically significant, suggesting that there is little evidence of UV
and blue reflectance being condition dependent, even when such analyses are
based on large sample sizes. Thus, there is little evidence of dorsal feather
coloration having evolved by the same selective forces that shaped other
secondary sexual characters in barn swallows.
A second way to investigate condition dependence of sexual signals is to
determine the relationship between expression of signals and body condition.
We found no evidence of significant correlations between plumage coloration
and body condition. This suggests that the structural coloration of the
plumage of the barn swallow has no clear condition dependence. This conclusion
contrasts with the situation in the blue grosbeak, where UV reflectance was
strongly positively correlated with body condition in a much smaller sample of
individuals (Keyser and Hill,
1999,
2000). Why characters with
structural colors should be condition dependent, and the mechanism underlying
such condition dependence, remains to be determined.
If the dorsal coloration had been a reliable condition-dependent signal, we
should have expected positive correlations with survival and age (review in
Jennions et al., 2001). If
there are clear sex differences in reflectance of the plumage, we should
expect selection for greater reflectance in one sex (males in the present
case) or selection against strong reflectance in the other sex (females).
However, no evidence was found for this prediction in barn swallows. When
considering viability selection in relation to reflectance, there was no
significant effect for either UV reflectance, UV chroma, or blue chroma.
Because breeding philopatry is extremely high in the barn swallow, with more
than 99% of all surviving adult birds returning to their previous breeding
site if alive (Møller,
1994b), we can conclude that there is little possibility for bias
in the data due to differences in recapture probability. This conclusion is
confirmed by mark—recapture analyses of the barn swallow data (de Lope
F, Møller AP, and Szép T, unpublished data). Thus, there is
little evidence of viability selection acting against females or males with
strong reflectance. These results contrast with findings in the bluethroat and
the blue tit Parus caeruleus indicating that older males have
stronger UV reflectance, or males with stronger reflectance survive better
(Andersson and Amundsen, 1997;
Sheldon et al., 1999).
Moreover, we found no evidence that color of an individual was maintained across year, nor was the ranking of individuals by their plumage coloration predictable across years. Those that showed a greater reflectance in 1999 on average did not show greater values in 2000. In conclusion, we could not find any clear evidence suggesting that reflectance of the iridescent plumage of male barn swallows is condition dependent.
Sexual selection and structural coloration
A number of studies has suggested that males with more extreme UV
reflectance enjoy a mating advantage compared to less ornamented males
(Andersson et al., 1998;
Bennett et al., 1996,
1997; Hunt et al.,
1997,
1999;
Johnsen et al., 1998). We
compared the phenotype of adult male barn swallows differing in mating status
but found no evidence that reflectance in the UV, UV chroma, or blue chroma
were associated with a mating advantage. Because males arriving on the same
date but differing in mating status were compared in this test, we were able
to control statistically for a number of confounding variables associated with
early arrival (Møller,
1994a). Such an approach has previously been used to show that
long-tailed male barn swallows enjoy a mating advantage
(Møller, 1990) and that
males with more intense red facial plumage coloration have greater mating
success (Ninni et al., in preparation). Thus, there was no direct evidence of
reflectance being associated with male mating success.
The duration of the premating period is a good measure of the
attractiveness of male barn swallows because it takes longer for less
attractive males to acquire a mate (Møller,
1988,
1990,
1994b). In the present study,
we found no significant relationship between the length of the premating
period and plumage coloration.
Male mating success in the barn swallow is associated with an early start
of breeding and hence with enhanced reproductive success (Møller,
1988,
1990,
1994b). UV reflectance of male
plumage was not significantly negatively correlated with reproductive success.
Although a negative correlation was found between UV chroma and blue chroma
with the number of fledglings in a season, given the number of pairs used in
the analysis (N = 56) and the number of tests performed in the study,
we do not consider this result to indicate a real cost for this character in
barn swallows. Finally, there was no significant correlation between
reflectance and parental care, a result that is similar to that of another
study by Smiseth et al.
(2001).
Previous studies of UV reflectance and extrapair paternity have shown a
direct, causal relationship (Johnsen et
al., 1998). We did not estimate extrapair paternity, which is an
important component of sexual selection in the barn swallow
(Ellegren et al., 1997;
Møller and Tegelström,
1997; Møller et al.,
1998; Saino et al.,
1997). However, previous studies of social mate choice and
extrapair paternity in the barn swallow have shown similar mate preferences at
work in the two contexts (Ellegren et al.,
1997; Møller and
Tegelström, 1997;
Møller et al., 1998;
Saino et al., 1997). Hence, we
suggest that it is unlikely that plumage reflectance will be significantly
related to extrapair paternity when it is not significantly related to a range
of other measures of sexual selection.
In conclusion, we found no evidence supporting the prediction that plumage reflectance should be directly associated with sexual selection.
Conclusions
The barn swallow has a number of different secondary sexual characters
including length and symmetry of the outermost tail feathers, the size of
white spots on the tail feathers, red facial plumage, and song rate. All these
traits are positively associated with male mating success. Models of multiple
sexual signals suggest that such traits will be indicators of different
aspects of condition, providing reliable information about the overall
condition of a signaler (review in
Møller and Pomiankowski,
1993). The presence of sexual dichromatism with respect to
reflectance in the UV and blue chroma, but neither of these being associated
with male mating success or condition, suggests that condition-dependent and
condition-independent traits may coexist in the same population of barn
swallows. Because UV reflectance and blue chroma are clearly sexually
dichromatic, we can infer that sexual selection must have acted on these
traits in the past. Given the observation that there is natural selection
against UV reflectance (Table
5), we can furthermore infer that sexual selection for increased
reflectance must occur regularly, although not in the present 2-year
study.
The presence of several dimorphic sexual traits in the same population
raises the question of which forces shaped the actual form and why this
difference is maintained. The coloration of the dorsal plumage seems not to
have any particular cost for barn swallows, except for its production, which
may be minor. The small difference in reflectance in the ultraviolet and the
lack of any dimorphism in UV chroma compared to the difference in blue chroma
supports the suggestion by Hunt et al.
(2001) that the ultraviolet
spectrum may not be a preferential channel for communication in birds. In barn
swallows it seems clear that other characters than dorsal plumage coloration
occur, with these characters being more costly to produce and maintain, and
thus more reliable indicators of condition.
|
ACKNOWLEDGEMENTS |
|---|
We are grateful to László Z. Garamszegi and all the people who helped with the field work and to J.-M. Rossi who provided invaluable help with software. We thank Claire Doutreland, Jesus Aviles, and three anonymous referees for constructive criticism that greatly improved the manuscript. F.d L. was supported by grants from the Spanish Ministry of Science and technology BOS2000-0293 and Junta de Extremadura IPR00A021.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Able KP, Able MA, 1993. Daytime calibration of magnetic orientation in a migratory bird requires a view of skylight polarization. Nature 364: 523-525.
Andersson M, 1994. Sexual selection. Princeton, New Jersey: Princeton University Press.
Andersson S, Amundsen T, 1997. Ultraviolet color
vision and ornamentation in bluethroats. Proc R Soc Lond B
264: 1587-1591.
Andersson S, Örnborg J, Andersson M, 1998.
Ultraviolet sexual dimorphism and assortative mating in blue tits. Proc
R Soc Lond B 265:
445-450.
Badyaev AV, Hill GE, Dunn PO, Glen JC, 2001. Plumage color as a composite trait: developmental and functional integration of sexual ornamentation. Am Nat 158: 221-235.[Web of Science][Medline]
Banks AN, 2001. For your eyes only? The role of UV in mate choice. Trends Ecol Evol 16: 473-474.
Bennett ATD, Cuthill IC, Norris KJ, 1994. Sexual selection and the mismeasure of color. Am Nat 144: 848-860.[Web of Science]
Bennett ATD, Cuthill IC, Partridge JC, Lunau K, 1997.
Ultraviolet plumage colors predict mate preferences in starlings. Proc
Natl Acad Sci USA 94:
8618-8621.
Bennett ATD, Cuthill IC, Partridge JC, Maier EJ, 1996. Ultraviolet vision and mate choice in zebra finches. Nature 380: 433-435.
Bowmaker JK, Heath LA, Wilkie SE, Hunt DM, 1997. Visual pigments and oil droplets from six classes of photoreceptors in the retinas of birds. Vision Res 37: 2183-2194.[Web of Science][Medline]
Camplani A, Saino N, Møller AP, 1999. Carotenoids, sexual signals and immune function in barn swallows from Chernobyl. Proc R Soc Lond B 266: 1111-1116.[Medline]
Chen D, Collins JS, Goldsmith TH, 1984. The UV
receptor of bird retinas. Science 225:
337-339.
Chen D, Goldsmith TH, 1986. Four spectral classes of cones in the retinas of birds. J Comp Physiol A 159: 473-479.[Medline]
Chandler CR, 1995. Practical considerations in the use of simultaneous inference for multiple tests. Anim Behav 49: 454-527.
Church SC, Bennet ATD, Church SC, Hart NS, Hunt S,
1998. Ultraviolet cues affect the foraging behaviour of blue
tits. Proc R Soc Lond B 265:
1509-1514.
Cuthill IC, Bennett ATD, Partridge JC, Maier EJ, 1999. Plumage reflectance and the objective assessment of avian sexual dichromatism. Am Nat 153: 183-200.[Web of Science]
Cuthill IC, Partridge JC, Bennett ATD, Church SC, Hart NS, Hunt S, 2000. Ultraviolet vision in birds. Adv Study Behav 29: 159-214.
de Lope F, Møller AP, 1993. Female reproductive effort depends on the degree of ornamentation of their mates. Evolution 47: 1152-1160.[Web of Science]
Ellegren H, Lindgren G, Primmer CR, Møller AP, 1997. Fitness loss and germline mutations in barn swallows breeding in Chernobyl. Nature 389: 593-596.[Medline]
Falconer DS, Mackay TFC, 1996. Introduction to quantitative genetics, 4th ed. New York: Longman.
Hunt S, Bennett ATD, Cuthill IC, Griffiths R, 1998.
Blue tits are ultraviolet tits. Proc R Soc Lond B
265: 451-455.
Hunt S, Cuthill IC, Bennet ATD, Church SC, Partridge JC, 2001. Is the ultraviolet wavelength a special communication channel in avian mate choice? J Avian Exp Biol 204: 2499-2507.
Hunt S, Cuthill IC, Bennett ATD, Griffiths R, 1999. Preferences for ultraviolet patterns in the blue tit. Anim Behav 58: 809-815.[Web of Science][Medline]
Hunt S, Cuthill IC, Swaddle JP, Bennett ATD, 1997. Ultraviolet vision and band-color preferences in female zebra finches, Taeniopygia guttata. Anim Behav 54: 1383-1392.[Web of Science][Medline]
Jacobs GH, 1992. UV vision in vertebrates. Am Zool 32: 544-554.
Jennions MD, Møller AP, Petrie M, 2001. Sexually selected traits and adult survival: a meta-analysis of the phenotypic relationship. Q Rev Biol 79: 3-36.
Johnsen A, Andersson S, Örnborg J, Lifjeld JT,
1998. Ultraviolet plumage ornamentation affects social mate
choice and sperm competition in bluethroats (Aves: Luscinia s.
svecica): a field experiment. Proc R Soc Lond B
265: 1313-1318.
Johnstone RA, 1995. Sexual selection, honest advertising and the handicap principle: reviewing the evidence. Biol Rev Camb Phil Soc 70: 1-65.[Medline]
Keyser AJ, Hill GE, 1999. Condition-dependent
variation in the blue-ultraviolet coloration of a structurally based plumage
ornament. Proc R Soc Lond B 266:
771-777.
Keyser AJ, Hill GE, 2000. Structurally based plumage
coloration is an honest signal of quality in male blue grosbeaks. Behav
Ecol 11:
202-209.
Kose M, Møller AP, 1998. Sexual selection, feather breakage and parasites: The importance of white spots in the tail of the barn swallow. Behav Ecol Sociobiol 45: 430-436.[Web of Science]
Møller AP, 1988. Female choice selects for male sexual tail ornaments in the monogamous swallow. Nature 322: 640-642.
Møller AP, 1989. Viability costs of male tail ornaments in a swallow. Nature 339: 132-135.
Møller AP, 1990. Male tail length and female mate choice in the monogamous swallow Hirundo rustica. Anim Behav 39: 458-465.[Web of Science]
Møller AP, 1992. Female preference for symmetrical male sexual ornaments. Nature 357: 238-240.[Medline]
Møller AP, 1993. Female preference for apparently symmetrical male sexual ornaments in the barn swallow Hirundo rustica. Behav Ecol Sociobiol 32: 371-376.[Web of Science]
Møller AP, 1994a. Phenotype-dependent arrival time and its consequences in a migratory bird. Behav Ecol Sociobiol 35: 115-122.[Web of Science]
Møller AP, 1994b. Sexual selection and the barn swallow. Oxford: Oxford University Press.
Møller AP, 1994c. Sexual selection in the barn swallow (Hirundo rustica). IV. Patterns of fluctuating asymmetry and selection against asymmetry. Evolution 48: 658-670.[Web of Science]
Møller AP, 1994d. Symmetrical male sexual
ornaments, paternal care, and offspring quality. Behav Ecol
5: 188-194.
Møller AP, 1995. Sexual selection in the barn swallow (Hirundo rustica). V. Geographic variation in ornament size. J Evol Biol 8: 3-19.
Møller AP, Pomiankowski A, 1993. Why have birds got multiple sexual ornaments? Behav Ecol Sociobiol 32: 167-176.[Web of Science]
Møller AP, Saino N, Taramino G, Galeotti P, Ferrario S, 1998. Paternity and multiple signaling: effects of a secondary sexual character and song on paternity in the barn swallow. Am Nat 151: 236-242.[Web of Science][Medline]
Møller AP, Tegelström H, 1997. Extra-pair paternity and tail ornamentation in the barn swallow Hirundo rustica. Behav Ecol Sociobiol 41: 353-360.[Web of Science]
Saino N, Primmer C, Ellegren H, Møller AP, 1997. An experimental study of paternity and tail ornamentation in the barn swallow (Hirundo rustica). Evolution 51: 562-570.[Web of Science]
Saino N, Stradi R, Ninni P, Møller AP, 1999. Carotenoid plasma concentration, immune profile and plumage ornamentation of male barn swallows (Hirundo rustica). Am Nat 154: 441-448.[Medline]
Sheldon BC, Andersson S, Griffiths S, Örnborg J, Sendecka J, 1999. Ultraviolet color variation influences blue tit sex ratios. Nature 402: 874-877.
Siitari H, Honkavaara J, Viitala J, 1999. Ultraviolet
reflection of berries attracts foraging birds. A laboratory study with
redwings (Turdus iliacus) and bilberries (Vaccinium
myrtillus). Proc R Soc Lond B 266:
2125-2129.
Smiseth PT, Örnborg J, Andersson S, Amundsen T,
2001. Is male plumage reflectance correlated with paternal care
in bluethroats? Behav Ecol 12:
164-170.
Stradi R, 1998. The color of flight. Milan: Solei Press.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. M Doucet and M. G Meadows Iridescence: a functional perspective J R Soc Interface, April 6, 2009; 6(Suppl_2): S115 - S132. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. P Veiga and V. Polo Feathers at nests are potential female signals in the spotless starling Biol Lett, September 22, 2005; 1(3): 334 - 337. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Z. Garamszegi, D. Heylen, A. P. Moller, M. Eens, and F. de Lope Age-dependent health status and song characteristics in the barn swallow Behav. Ecol., May 1, 2005; 16(3): 580 - 591. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||